Map&Reduce Verfahren

/*
 * 
 * "Information Systems"
 *
 * CloudClust
 */

package de.unidue.inf.is.cloudclust.server.model.labeling;

import java.util.ArrayList;
import java.util.List;

/**
 * 
 * @author XYZ
 *
 */
public class ClusterLabel {

    private ArrayList<Double> tf = new ArrayList<Double>();
//    private ArrayList<Double> idf = new ArrayList<Double>();
    private ArrayList<String> terms = new ArrayList<String>();
    private ArrayList<String> clusterId = new ArrayList<String>();
//    private ArrayList<String> dokVecId = new ArrayList<String>();

    private ArrayList<Double> tfmean = new ArrayList<Double>();
    private ArrayList<Double> tfmeanclone = new ArrayList<Double>();
    private ArrayList<String> termssort = new ArrayList<String>();
    private ArrayList<String> firstterms = new ArrayList<String>();

    public List<String> getLabel() {
        //Method getDocVecCollecion
        //write ArrayList 

        double h = 0;
        int anzahl = 0;
        for (String term : terms) {
            int pos = terms.indexOf(term);
            for (String dok : clusterId) {
                //created relation between tf and document
                h = h + tf.get(pos);
                anzahl++;
            }
            tfmean.add(pos, (h / anzahl));
        }

        //start sorting here
        for (Double wert : tfmean) {
            tfmeanclone.add(wert);
        }
        double max = 0;
//        int index;
        while (tfmeanclone.size() != 0) {
            for (double wert : tfmeanclone) {
                if (wert > max) {
                    max = wert;
                }
            }
            //check if same index
            termssort.add(terms.get(tfmeanclone.indexOf(max)));
            tfmeanclone.remove(tfmeanclone.indexOf(max));
        }
        return termssort;
    }



public List<String> getFirst(){
    int x=0;
    while (x <=20){
    
    firstterms.add(termssort.get(x));
    x++;
    }
     return firstterms;
  }
}